The proposed research is a longitudinal and sequential follow-up of a group of patients with prior acute coronary heart attacks. The overall goal is to develop and validate a method which will prospectively identify and characterize those individuals who die suddenly (abruptly or within 24 hours after symptom onset) from a repeat acute coronary heart attack. Prospective data to be collected include details of the severity and type of coronary heart attack at entrance into the study and triannual follow-up information of 6-hour taped ambulatory cardiac rhythm recordings, interim symptoms and affect state. These variables will be related to a variety of clearly defined endpoints including the severity of repeated coronary heart attacks, sudden cardiac death and severity of follow-up cardiac arrhythmias. Data processing techniques will be utilized for data handling, storage and retrieval and data analysis which involves both Chi squared tabular relationships and step-wise discriminant analysis techniques.